Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks51 from public data
- MotionBERT: A Unified Perspective on Learning Human Motion Representations428
A pretraining stage in which a motion encoder is trained to recover the underlying 3D motion from noisy partial 2D observations is proposed, which achieves state-of-the-art performance on all three downstream tasks by simply finetuning the pretrained motion encoding with a simple regression head, which demonstrates the versatility of the learned motion representations.
- TAM: Temporal Adaptive Module for Video Recognition388
A new temporal adaptive module (TAM) to generate video-specific temporal kernels based on its own feature map that outperforms other temporal modeling methods consistently, and achieves the state-of-the-art performance under the similar complexity.
- TEINet: Towards an Efficient Architecture for Video Recognition275
The proposed TEINet can achieve a good recognition accuracy on these datasets but still preserve a high efficiency, and is able to capture temporal structure flexibly and effectively, but also efficient for model inference.
- InternGPT: Solving Vision-Centric Tasks by Interacting with ChatGPT Beyond Language113
By incorporating pointing instructions, the proposed iGPT significantly improves the efficiency of communication between users and chatbots, as well as the accuracy of chatbots in vision-centric tasks, especially in complicated visual scenarios where the number of objects is greater than 2.
- Dynamic Sampling Networks for Efficient Action Recognition in Videos87
The experimental results demonstrate that DSN is able to greatly improve the inference efficiency by only using less than half of the clips, which can still obtain a slightly better or comparable recognition accuracy to the state-of-the-art approaches.
- What is Wrong with Perplexity for Long-context Language Modeling?77
Deeper insights are offered into the limitations of PPL and effective solutions for accurately evaluating and enhancing the long-context capabilities of LLMs are presented, leading to consistent improvements across diverse benchmarks.
- VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling61
This work presents a large-scale dataset by collecting 360K video-music pairs, and proposes VidMuse, a simple framework for generating music aligned with video inputs that outperforms existing models in terms of audio quality, diversity, and audio-visual alignment.
- ZeroGUI: Automating Online GUI Learning at Zero Human Cost38
ZeroGUI is proposed, a scalable, online learning framework for automating GUI Agent training at Zero human cost that integrates VLM-based automatic task generation to produce diverse training goals from the current environment state, and two-stage online reinforcement learning to continuously interact with and learn from GUI environments.
- 36
- 33
- 29
- Progressive Attention on Multi-Level Dense Difference Maps for Generic Event Boundary Detection28
An effective and end-to-end learnable framework (DDM-Net) to tackle the diversity and complicated semantics of event boundaries, and demonstrates the effectiveness of richer motion representation and more sophisticated aggregation, in handling the diversity of GEBD.
- 23
- MMTrail: A Multimodal Trailer Video Dataset with Language and Music Descriptions18
This work presents MMTrail, a large-scale multi-modality video-language dataset incorporating more than 20M trailer clips with visual captions, and 2M high-quality clips with multimodal captions incorporating various modality annotations, and proposes a systemic captioning framework.
- Medical Image Encryption using Biometric Image Texture Fusion14
A new texture fusion medical image encryption (TFMIE) algorithm derived from biometric images is proposed, which can encrypt the image using biometric information for storage or transmission and has high transmission efficiency.
- Visual image encryption based on compressed sensing and Cycle-GAN11
A learning visual image encryption scheme based on compressed sensing and cycle generative adversarial network that has sufficient key space, strong key sensitivity, and high reconstruction quality is proposed.
- Multi-agent generalized cooperative optimization scheduling for multi-energy complementarity in microgrids8
A novel multi-objective generalized normal distribution optimization (MGNDO) algorithm is proposed to enhance the correlation of policy sampling, and by updating the covariance matrix, the policy correlations between different agents are better captured, resulting in more cooperative action sequences.
- 6
- 6
- VLG: General Video Recognition with Web Textual Knowledge5
A unified visual-linguistic framework (VLG) to solve the problem of GVR with an effective two-stage training paradigm is presented, and extensive results show that the VLG obtains the state-of-the-art performance under four settings.
- Part-and-whole: A novel framework for deformable medical image registration4
A novel part-and-whole registration framework for 3D medical images consisting of two encoder-decoder architectures which take image pairs and patch pairs as inputs, respectively, and estimate the deformation field as the result and shows that the method consistently outperforms recent state-of-the-art baselines.
- Adaptive Image Restoration via a Relaxed Regularization of Mean Curvature4
A new relaxation model based on mean curvature for adaptive image restoration based on an alternating direction method of multipliers (ADMMs) is proposed and a rigorous convergence theory of the proposed algorithm is established.
- Visually meaningful image encryption combining with reversible data hiding in encrypted images3
A visually meaningful image encryption (VMIE) algorithm combining with Reversible Data Hiding in Encrypted Images (RDHEI) is proposed, which combines with the RDHEI, improving the transmission efficiency and has no image quality problems.
- 3
- 3
- 2
- 2
- The EEG Signal Encryption Scheme Based on Hyperchaotic System and Identity Authentication2
An EEG signal encryption scheme that combines the hyperchaotic system and identity authentication is proposed and the experimental results show that the encryption scheme has high security, strong key sensitivity and anti-attack capability.
- 1
- An All-Optical Metro Network Architecture and QoS-Aware Wavelength Allocation Study for Converged Fixed, Mobile, and Edge Computing Multi-Granular Traffic1
An all-optical metro network architecture, called MOON, to serve converged multigranular traffic from fixed, mobile, and edge computing services, which incorporates hybrid optical switching (HOS) combining optical circuit switching (OCS) and optical time slotted switching (OTS) capabilities.
- 1
- Lightweight Image Super-Resolution Reconstruction Guided by Gated Feedback and LatticeFormer1
A lightweight image super-resolution reconstruction model guided by gated feedback and LatticeFormer that can generate high-quality reconstructed images with complete structures and clear texture details is proposed.
- 1
- 1
- 1
- 1
- Dynamic Secure Key Distribution Based on Dispersion Equalization and Cellular Automata for Optical Transmission1
The scheme effectively eliminates the key inconsistency problem caused by imperfect channel reciprocity, and dynamic key sequences can be conveniently generated with large key space in long-haul optical transmission.
- Submission to Generic Event Boundary Detection Challenge@CVPR 2022: Local Context Modeling and Global Boundary Decoding Approach1
A local context modeling sub-network is proposed to perceive diverse patterns of generic event boundaries, and it generates powerful video representations and reliable boundary confidence, and global boundary decoding sub- network is exploited to decode event boundaries from a global view.
- –
- –
- –
- –
- A Deep Learning-based Robotic System for Fully Automated Milk Tea Preparation–
This research introduces a fully integrated system that combines deep learning-based visual perception with a simplified robotic control strategy that significantly reduces computational complexity while maintaining high reliability, offering a practical solution for deploying intelligent robotic systems in real-world food preparation scenarios with limited resources.
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.